Re: Lilypond 2.19.82 with multiple voices seems to fail

[Phil Holmes]

"DaveAtkinson" <address@hidden> wrote in message 
news:address@hidden
> Dear LilyPond folks
>
> Thanks so much for taking an interest. First, I forgot two important 
> pieces
> of information - yes, I am using Windows... (hangs head in shame) and it 
> has
> 16 Gb of memory.. The OS is 64 bit, but I don't suppose that matters to
> lilypond as compiled for windows..?
>
> So interesting... it is clear that overuse of the << \\ >> construct 
> causes
> issues. I would welcome comments about how to re-write the score without 
> it.
> If the correct way to do it is
>
>
> then an affirmation would be sufficient rather than a tutorial, which 
> would
> be greatly appreciated otherwise ;-).

I can't see your example, but I would do your first two bars like this:

global = {
 \key f \major
 \time 3/4
}

rhnotesupper = \relative c' {
 \clef treble
 \global

 \tempo 4=83

 s4 s <f a>16-.( <f a>-. <f a>-. <f a>-.) |
 \voiceOne <f' a>-.( <f a>-. <f a>-. <f a>-.) \oneVoice <f, a>16-.( <f a>-. 
<f a>-. <f a>-.) <f, a>-.( <f a>-. <f a>-. <f a>-.) |
}

rhnoteslower = \relative c' {
 \clef treble
 \global

 \tempo 4=83

 s1*3/4 |
 \voiceTwo <f c'>4 s s |
}

lhnotesupper =  \relative c {
 \clef bass
 \global

 \voiceOne <a c f>16-.(  <a c f>-.  <a c f>-.  <a c f>-.) <f' a>-.( <f a>-. 
<f a>-. <f a>-.) \oneVoice <f c'>4 |
 s <f c'> <f, c'> |
}

lhnoteslower =  \relative c {
 \clef bass
 \global

 \voiceTwo f,,4\sustainOn <f' c'> |
 s1*3/4 |
}

\score {
 \new PianoStaff \with { instrumentName = #"Piano" }
 <<
   \new Staff = "upper"
   <<
     \new Voice = "top" \rhnotesupper
     \new Voice = "next" \rhnoteslower
   >>
   \new Staff = "lower"
   <<
     \new Voice = "topLH" \lhnotesupper
     \new Voice = "nextLH" \lhnoteslower
   >>
 >>
 \layout { }
 \midi { }
}

> The first, I confess to sometimes using lilypond to generate midi files 
> from
> piano accompaniments so that I can practise the main part when I don't 
> have
> the luxury of a pianist (which is almost all the time). I am therefore
> copying existing scores. In order to eliminate (!) mistakes, I do it a few
> bars at a time, as corrections are much easier for me to do in the last 
> few
> bars. If I were to use two voices for the whole of a staff, once I'd 
> written
> the first staff, going back and finding the places where two voices are
> required, and filing in all the spaces, is tedious and taxing for me...

I create music (normally from renaissance polyphony) and simply have four 
voices and add a few bars to each at a time.  I then compile, check and add 
a few more bars to each.  My only other suggestion (which Kieren, a regular 
contributor and composer would agree with) is that you might find not using 
\relative means it's easier to keep track of which octave you're currently 
in.

> The second reason, if I were to write a staff as two lines, all the parts
> where there is effectively only a single voice look horrible, because all
> the rests and note stems look wrong.

\oneVoice, \voiceOne, \voiceTwo etc. are your friends.

--
Phil Holmes